arXiv subjects

Junjie Zhu

Publications and source records attributed to Junjie Zhu.

At least 19 recordsLinked to original sources

arXiv2026

Surface Commissioning and Performance of the sMDT Muon Chambers for the ATLAS HL-LHC Upgrade

To improve the first-level muon trigger efficiency at the HL-LHC, the Monitored Drift Tube (MDT) chambers in the inner small sectors of the ATLAS Barrel Muon Spectrometer will be replaced with integrated modules of small-Diameter Muon Drift Tube (sMDT) and Resistive Plate Chambers (RPCs). This paper reports on the surface commissioning of 102 new sMDT chambers at CERN in 2025 following the installation of the final front-end electronics, including the new Amplifier-Shaper-Discriminator (ASD), Time-to-Digital Converter (TDC) chips, and Chamber Service Module (CSM) developed for the HL-LHC. The commissioning included measurements of gas leak rates and high-voltage dark currents, tests of the integrated planarity monitoring system, noise characterization, and measurements of detector efficiency and spatial resolution using cosmic rays. Fewer than 0.1\% of the 49152 tubes were found to be non-functional due to broken sense wires or gas leaks. The chambers exceed the ATLAS design requirements, with gas leak rates a factor of five below the specified limit of $9.3\times 10^{-3}~\mathrm{mbar\cdot liter/s}$ per chamber, dark currents of only around 0.2~nA per tube, average channel noise hit rates below 30~Hz, and average drift tube efficiency and spatial resolution of 99\% and $82~μ\mathrm{m}$, respectively, at an effective threshold of 15 primary electrons.

physics.ins-det↗

arXiv2026

High-Speed Semi-FE Readout Module for ATLAS MDT at HL-LHC: Design and Production-Level Characterization

The High-Luminosity upgrade of the Large Hadron Collider (HL-LHC) introduces increased demands on the ATLAS Muon Spectrometer, particularly in terms of data throughput, timing distribution and system reliability. The Phase-II Chamber Service Module (CSM) is a key component of the upgraded Monitored Drift Tube (MDT) trigger and readout system, providing a high-speed interface between the front-end electronics and the backend systems. This paper describes the design and implementation of the Phase-II CSM, together with its validation. The results show that the CSM supports two independent optical uplinks, each operating at a line rate of 10.24 Gbps, together with clock distribution and slow control in the expected operating environment. Integration with small-diameter MDT (sMDT) chambers and tests with the prototype L0MDT trigger system are also presented. The CSM boards are now in production and will be used for installation and integration during the upcoming LHC Long Shutdown.

physics.ins-det↗

arXiv2026

Tactile-WAM: Touch-Aware World Action Model with Tactile Asymmetric Attention

World Action Models (WAMs) jointly predict future visual observations and actions, but visual futures alone often miss slip, jamming, contact-direction changes, and subtle misalign- ment in contact-rich manipulation. Tactile signals reveal these hidden physical states, yet naive tactile-token injection can disrupt visual dynamics modeling due to the limited scale of tactile data, a phenomenon we term tactile pollution. We in- troduce Tactile-WAM, which uses asymmetric attention to block video queries from tactile keys while preserving tac- tile access for action queries. A contact-change-aware bias further strengthens action attention to touch. Because tactile pixel changes do not reliably reflect contact changes, we derive Observed proxy changes drive the attention bias, while future- proxy supervision preserves action-relevant contact dynamics in predicted tactile representations. On ManiFeel, visual-path isolation reduces deviation from the RGB-only trajectory by 21.8% in MSE at the step-matched 20K checkpoint without a statistically detectable change in ground-truth video qual- ity. The full model improves average success from 15.6% to 32.7%, with VideoClean providing the largest gain. On five real-robot tasks, Tactile-WAM achieves 49.2% success.

cs.RO↗

arXiv2026

PROOF-Gen: From Optimized Data to Better Distillation

Supervised fine-tuning on teacher-generated trajectories is the standard first stage for distilling tool-calling capabilities into deployable models. Post-training pipelines that drive shipped tool-calling agents re-run this stage on a daily or weekly cadence, paying the frontier-teacher cost each cycle, yet the mechanism is generate-and-filter (keep the teacher's passing trajectories, discard the rest) and each cycle leaves behind the same hard scenarios because failures supply no signal. On τ2-bench, 57% of teacher trials fail, two-thirds of them near-misses (most tool calls correct, undone by one decisive error). We introduce PROOF-Gen (Per-scenario Reflective Optimization to Overcome FailedGeneration), which recovers golden trajectories from these failures via per-scenario prompt optimization. For each failed task, a reflector analyzes the execution trace and evaluation feedback, then writes corrective guidance that steers the teacher to a passing trajectory. The guidance is stripped before training, so the student learns from clean demonstrations with no task-specific scaffold. On τ2-bench, per-scenario optimization recovers 93% of failed scenarios. Fine-tuned on the combined data, Qwen3-4B-Instruct-2507 improves from Pass^1=0.132 to 0.529 and Gemma 4 E4B-it gains +7.2pp on BFCL v4 multi-turn. In a deployed pipeline, the method lifts trajectory quality by +6.3pp goal completion and transfers to a deployed on-device model (+1.5pp goal completion; +1.7 to +5.0pp across response-quality metrics), with positive transfer in every locale (non-English average +1.48pp).

cs.AI↗

arXiv2026

GST-Bench: Can VLMs Develop Global Spatial Awareness from Video?

Spatial intelligence is fundamental to embodied agents, yet existing benchmarks focus on local spatial perception from single or few viewpoints, overlooking global spatial awareness over continuous, long-horizon visual streams. To address this limitation, we introduce the Global-Spatial-Temporal Benchmark (GST-Bench), a VQA benchmark for global spatial intelligence in video understanding, comprising human-verified questions derived from 6,790 minutes of synthetically generated video. It requires models to perform accurate spatial inference from novel viewpoints unseen in the input video and to map egocentric observations onto global top-down images. A comprehensive evaluation of 22 state-of-the-art VLMs exposes a striking gap between models and humans: the strongest zero-shot model attains only 42.68, far below the human score of 79.08. To probe the cause of this gap, we construct GST-Bench-Local and find that models, despite strong local spatial understanding under the same task formulation, still fail to consolidate long-horizon observations into a globally consistent scene representation. We further provide GST-Train, a dataset for global spatial reasoning, as a complementary resource to facilitate future research on this challenge.

cs.CV↗

arXiv2026

Not All Patches are Equal: Sampling Matters for Visible-Infrared Pre-Training

Visible-infrared (VIS-IR) alignment is a key pre-training task for robust multi-sensor perception. Most existing methods use uniform patch-wise contrastive learning, but this can be unreliable in VIS-IR data because imaging-physics differences make some spatially paired regions inherently less comparable, and aligning them with equal strength hinders representation learning and downstream transfer. In this paper, we revisit VIS-IR pre-training from a sampling perspective and propose Importance-Aware Sampling (IAS), which adjusts training emphasis based on patch reliability. Specifically, IAS (i) derives patch weights from infrared structural cues and uses them to reweight the contrastive objective; (ii) learns a soft importance mask with a lightweight sampler, optionally warm-started from the hand-crafted prior; and (iii) employs a patch curriculum learning strategy that gradually expands from high-reliability regions to harder patches. It is worth noting that IAS is plug-and-play and works with both patch-/correlation-level alignment (e.g., UNIV-style) and image-level contrastive baselines (e.g., ImageBind-style). Extensive experiments on multiple VIS-IR benchmarks demonstrate consistent improvements over strong baselines, including for IR semantic segmentation, IR object detection and VIS semantic segmentation and cross-modal retrieval task. Code will be released on https://github.com/KlayMa527/IAS.

cs.CV↗

arXiv2026

A 24-Channel Ultra-Low-Noise Preamplifier for dN/dx Measurements with Drift Tube Detectors

Cluster counting dN/dx is a promising method to enhance particle identification for gaseous detectors, especially in next-generation collider experiments like the FCC-ee, where good pion-kaon separation over a broad momentum range is essential. However, its implementation in large-scale systems has been limited by the challenging requirements for high-resolution signal amplification and readout. This paper presents a 24-channel ultra-low-noise preamplifier board designed for drift tube detectors to enable dN/dx measurements. The three-stage amplification topology employs SiGe transistors and integrates dedicated noise-minimization techniques, achieving a charge gain of 21.11 mV/fC from 0.3 fC to 50 fC, a bandwidth of 542 MHz, and a voltage gain of 47.8 dB. The measured voltage noise density is 0.35 nV/sqrt(Hz), surpassing most of the state-of-the-art preamplifiers for gaseous and silicon detectors. Validation tests conducted on the sMDT chambers at the CERN Proton Synchrotron test beam facility demonstrate that the proposed design meets the stringent preamplifier requirements for implementing the dN/dx method in drift-tube detector systems, achieving an equivalent noise charge of 0.14 fC and a signal-to-noise ratio of 73 when operated with a He:iC4H10 (90:10) gas mixture. The design also shows promise for broader application in other gaseous or semiconductor detectors.

physics.ins-det↗

arXiv2026

Reinforced Agent: Inference-Time Feedback for Tool-Calling Agents

Tool-calling agents are evaluated on tool selection, parameter accuracy, and scope recognition, yet LLM trajectory assessments remain inherently post-hoc. Disconnected from the active execution loop, such assessments identify errors that are usually addressed through prompt-tuning or retraining, and fundamentally cannot course-correct the agent in real time. To close this gap, we move evaluation into the execution loop at inference time: a specialized reviewer agent evaluates provisional tool calls prior to execution, shifting the paradigm from post-hoc recovery to proactive evaluation and error mitigation. In practice, this architecture establishes a clear separation of concerns between the primary execution agent and a secondary review agent. As with any multi-agent system, the reviewer can introduce new errors while correcting others, yet no prior work to our knowledge has systematically measured this tradeoff. To quantify this tradeoff, we introduce Helpfulness-Harmfulness metrics: helpfulness measures the percentage of base agent errors that feedback corrects; harmfulness measures the percentage of correct responses that feedback degrades. These metrics directly inform reviewer design by revealing whether a given model or prompt provides net positive value. We evaluate our approach on BFCL (single-turn) and Tau2-Bench (multi-turn stateful scenarios), achieving +5.5% on irrelevance detection and +7.1% on multi-turn tasks. Our metrics reveal that reviewer model choice is critical: the reasoning model o3-mini achieves a 3:1 benefit-to-risk ratio versus 2.1:1 for GPT-4o. Automated prompt optimization via GEPA provides an additional +1.5-2.8%. Together, these results demonstrate a core advantage of separating execution and review: the reviewer can be systematically improved through model selection and prompt optimization, without retraining the base agent.

cs.AI↗

arXiv2026

Performance Evaluation of Straw Tubes with Muon Beams at CERN

We present results from two test beam campaigns that investigate the performance of straw tube detectors as potential candidates for an FCC-ee straw tracker. These studies were carried out at CERN using 150 GeV muon beams. Dedicated algorithms were developed to determine both single tube spatial resolution for the primary coordinate in the $r-ϕ$ plane and spatial resolution for the secondary coordinate along the tube direction within a straw chamber. Detection efficiency was also evaluated as a function of the extrapolated hit position for each tube. Both datasets showed consistent results for spatial resolutions and efficiency. Our findings will help establish benchmark performance metrics and provide valuable insight for future design, optimization, and construction of straw chambers for high-precision tracking applications.

physics.ins-det↗

arXiv2026

A Utility-preserving De-identification Pipeline for Cross-hospital Radiology Data Sharing

Large-scale radiology data are critical for developing robust medical AI systems. However, sharing such data across hospitals remains heavily constrained by privacy concerns. Existing de-identification research in radiology mainly focus on removing identifiable information to enable compliant data release. Yet whether de-identified radiology data can still preserve sufficient utility for large-scale vision-language model training and cross-hospital transfer remains underexplored. In this paper, we introduce a utility-preserving de-identification pipeline (UPDP) for cross-hospital radiology data sharing. Specifically, we compile a blacklist of privacy-sensitive terms and a whitelist of pathology-related terms. For radiology images, we use a generative filtering mechanism that synthesis a privacy-filtered and pathology-reserved counterparts of the original images. These synthetic image counterparts, together with ID-filtered reports, can then be securely shared across hospitals for downstream model development and evaluation. Experiments on public chest X-ray benchmarks demonstrate that our method effectively removes privacy-sensitive information while preserving diagnostically relevant pathology cues. Models trained on the de-identified data maintain competitive diagnostic accuracy compared with those trained on the original data, while exhibiting a marked decline in identity-related accuracy, confirming effective privacy protection. In the cross-hospital setting, we further show that de-identified data can be combined with local data to yield better performance.

cs.CV↗

arXiv2026

ATLAS MDT TDC Simulations for LHC Run3 and HL-LHC

The Large Hadron Collider (LHC) started the Run 3 operation in 2022, and the peak instantaneous luminosity in Run 3 may reach 3 x 10^34 cm-2s-1. The ATLAS Monitored Drift Tube (MDT) chambers are the main component of the precision tracking system in the ATLAS muon spectrometer. It is important to understand any potential issues with the MDT Front-End (FE) readout electronics for an expected level-1 (L1) trigger rate of 100 kHz and a complex deadtime of over 5% for Run 3 operations. We use raw data collected in 2022 to emulate the expected hit rates in MDT chambers and perform a realistic simulation on the ATLAS Muon TDC (Time-to-Digital Converter) (AMT) chip with the current configuration. We study the AMT chip performances by analyzing the trigger/L1/readout buffer occupancies and hit loss fractions under different luminosities with L1 rate of 100 kHz by using the Modelsim software. The hit loss fraction of the hottest MDT chamber (BIL3C05) is lower than 5% due to FE readout, even at a luminosity of 5.01 x 10^34 cm-2s-1 with a deadtime of 5% and a L1 rate of 100 kHz, indicating that AMT can operate under Run 3 conditions without problems. The MDT trigger and readout electronics will be replaced for triggerless readout during High-Luminosity LHC (HL-LHC) runs. We also simulate the AMT behavior in the triggerless mode up to 7.44 x 10^34 cm-2s-1 and propose possible AMT configurations in case some FE electronics could not be replaced during the long shutdown 3 (LS3).

physics.ins-det↗

arXiv2025

HyperLoad: A Cross-Modality Enhanced Large Language Model-Based Framework for Green Data Center Cooling Load Prediction

The rapid growth of artificial intelligence is exponentially escalating computational demand, inflating data center energy use and carbon emissions, and spurring rapid deployment of green data centers to relieve resource and environmental stress. Achieving sub-minute orchestration of renewables, storage, and loads, while minimizing PUE and lifecycle carbon intensity, hinges on accurate load forecasting. However, existing methods struggle to address small-sample scenarios caused by cold start, load distortion, multi-source data fragmentation, and distribution shifts in green data centers. We introduce HyperLoad, a cross-modality framework that exploits pre-trained large language models (LLMs) to overcome data scarcity. In the Cross-Modality Knowledge Alignment phase, textual priors and time-series data are mapped to a common latent space, maximizing the utility of prior knowledge. In the Multi-Scale Feature Modeling phase, domain-aligned priors are injected through adaptive prefix-tuning, enabling rapid scenario adaptation, while an Enhanced Global Interaction Attention mechanism captures cross-device temporal dependencies. The public DCData dataset is released for benchmarking. Under both data sufficient and data scarce settings, HyperLoad consistently surpasses state-of-the-art (SOTA) baselines, demonstrating its practicality for sustainable green data center management.

cs.LG↗

arXiv2025

RLHFSpec: Breaking the Efficiency Bottleneck in RLHF Training via Adaptive Drafting

Reinforcement Learning from Human Feedback (RLHF) is an important fine-tuning technique for large language models (LLMs) and comprises three stages: generation, inference, and training. The generation stage generates samples that are then used to infer learnable experiences for training. We observe that the generation stage is the bottleneck of the entire execution process and consider it a key point for optimization. Specifically, we realize the first attempt to integrate speculative decoding into the RLHF generation stage and propose RLHFSpec, an RLHF system that accelerates generation execution with efficient speculative decoding and sample reallocation. To fully exploit the performance potential provided by speculative decoding, especially dealing with the dynamic workload of the generation stage, RLHFSpec proposes a workload-aware drafting strategy selection mechanism, which selects the near-optimal strategy by jointly considering the verification cost and the number of accepted tokens. Moreover, RLHFSpec also proposes sample reallocation to fully utilize the GPU resources, and optimizes it with an efficient sample migration mechanism. The experimental results show that the RLHFSpec can achieve higher throughput in the generation stage compared to state-of-the-art works. Moreover, due to the effective alleviation of the generation bottleneck, RLHFSpec also shows significant performance speedup in the entire RLHF execution.

cs.LG↗

arXiv2025

SAR-TEXT: A Large-Scale SAR Image-Text Dataset Built with SAR-Narrator and A Progressive Learning Strategy for Downstream Tasks

Vision Language Models (VLMs) have achieved remarkable breakthroughs in the field of remote sensing in recent years. Synthetic Aperture Radar (SAR) imagery, with its all-weather capability, is essential in remote sensing, yet the lack of large-scale, high-quality SAR image-text datasets hinders its semantic understanding. In this paper, we construct SAR-TEXT, a large-scale and high-quality dataset consisting of over 130,000 SAR image-text pairs. To construct the SAR-TEXT dataset, we design the SAR-Narrator framework, which generates textual descriptions for SAR images through a multi-stage strategy. To verify the effectiveness of the SAR-TEXT dataset, we conduct experiments on three typical vision-language tasks: image-text retrieval, image captioning, and visual question answering (VQA). Specifically, we construct three representative models on SAR-TEXT: SAR-RS-CLIP, SAR-RS-CoCa, and SAR-GPT. SAR-RS-CLIP achieves notable improvements in retrieval performance, boosting average recall by 12.97% and 10.0% on the OSdataset_512 and HRSID test sets, respectively. In the captioning task, SAR-RS-CoCa achieves significant improvements over the original CoCa models in terms of BLEU-4, SPICE, and CIDEr scores. In the VQA task, SAR-GPT outperforms baseline and single-stage models on multiple SAR-VQA datasets, demonstrating stronger semantic understanding and reasoning ability, as further confirmed by qualitative results. It is worth noting that, as a flexible captioning tool, SAR-Narrator can be readily adopted by the community to construct larger-scale SAR image-text datasets. All code, pretrained models, and the SAR-Text dataset are publicly available at: https://github.com/YiguoHe/SAR-TEXT.

cs.CV↗

arXiv2025

Enhancing Remote Sensing Vision-Language Models Through MLLM and LLM-Based High-Quality Image-Text Dataset Generation

The application of Vision-language foundation models (VLFMs) to remote sensing (RS) imagery has garnered significant attention due to their superior capability in various downstream tasks. A key challenge lies in the scarcity of high-quality, large-scale, image-text paired training data. Recently, several works introduced extensive image-text datasets for RS and trained their VLFMs. However, due to the rudimentary methods used for generating captions, the quality of datasets is suboptimal, requiring larger volumes of training data, while only yielding modest performance improvements. In this paper, we propose a two-stage method named MpGI(Multi-Perspective Generation and Integration) for generating high-quality text captions for RS images. Firstly, we generate distinct and detailed descriptions from different perspectives using Rule-MLLM(Multimodal Large Language Model) Relay Generation and MLLMs generation methods. Next, we utilize Large Language Models (LLMs) to integrate these diverse descriptions into comprehensive captions, capturing details from multiple perspectives. Finally, we have created the HQRS-IT-210K dataset, including about 210,000 RS images and 1.3 million captions. We fine-tuned two VLFMs using our dataset: CLIP, a discriminative model, and CoCa, an image-to-text generative model. This process resulted in our proposed HQRS-CLIP and RS-CoCa models. Experimental results demonstrate that HQRS-CLIP surpassed the previous SOTA RS CLIP model in various downstream tasks while using only 4.2\% of the training data. RS-CoCa outperforms other advanced approaches across benchmark datasets and can generate captions for RS images that rival or even exceed manual annotations. Dataset, pre-trained models, and codes will be released at https://github.com/YiguoHe/HQRS-210K-and-HQRS-CLIP.

cs.CV↗

arXiv2025

Bridging VLM and KMP: Enabling Fine-grained robotic manipulation via Semantic Keypoints Representation

From early Movement Primitive (MP) techniques to modern Vision-Language Models (VLMs), autonomous manipulation has remained a pivotal topic in robotics. As two extremes, VLM-based methods emphasize zero-shot and adaptive manipulation but struggle with fine-grained planning. In contrast, MP-based approaches excel in precise trajectory generalization but lack decision-making ability. To leverage the strengths of the two frameworks, we propose VL-MP, which integrates VLM with Kernelized Movement Primitives (KMP) via a low-distortion decision information transfer bridge, enabling fine-grained robotic manipulation under ambiguous situations. One key of VL-MP is the accurate representation of task decision parameters through semantic keypoints constraints, leading to more precise task parameter generation. Additionally, we introduce a local trajectory feature-enhanced KMP to support VL-MP, thereby achieving shape preservation for complex trajectories. Extensive experiments conducted in complex real-world environments validate the effectiveness of VL-MP for adaptive and fine-grained manipulation.

cs.RO↗

arXiv2025

Graph-Based Cross-Domain Knowledge Distillation for Cross-Dataset Text-to-Image Person Retrieval

Video surveillance systems are crucial components for ensuring public safety and management in smart city. As a fundamental task in video surveillance, text-to-image person retrieval aims to retrieve the target person from an image gallery that best matches the given text description. Most existing text-to-image person retrieval methods are trained in a supervised manner that requires sufficient labeled data in the target domain. However, it is common in practice that only unlabeled data is available in the target domain due to the difficulty and cost of data annotation, which limits the generalization of existing methods in practical application scenarios. To address this issue, we propose a novel unsupervised domain adaptation method, termed Graph-Based Cross-Domain Knowledge Distillation (GCKD), to learn the cross-modal feature representation for text-to-image person retrieval in a cross-dataset scenario. The proposed GCKD method consists of two main components. Firstly, a graph-based multi-modal propagation module is designed to bridge the cross-domain correlation among the visual and textual samples. Secondly, a contrastive momentum knowledge distillation module is proposed to learn the cross-modal feature representation using the online knowledge distillation strategy. By jointly optimizing the two modules, the proposed method is able to achieve efficient performance for cross-dataset text-to-image person retrieval. acExtensive experiments on three publicly available text-to-image person retrieval datasets demonstrate the effectiveness of the proposed GCKD method, which consistently outperforms the state-of-the-art baselines.

cs.CV↗

arXiv2024

Fourier dimension of constant rank hypersurfaces

Any hypersurface in $\mathbb{R}^{d+1}$ has a Hausdorff dimension of $d$. However, the Fourier dimension depends on the finer geometric properties of the hypersurface. For example, the Fourier dimension of a hyperplane is 0, and the Fourier dimension of a hypersurface with non-vanishing Gaussian curvature is $d$. Recently, Harris showed that the Euclidean light cone in $\mathbb{R}^{d+1}$ has a Fourier dimension of $d-1$, which leads one to conjecture that the Fourier dimension of a hypersurface equals the number of non-vanishing principal curvatures. We prove this conjecture for all constant rank hypersurfaces. Our method involves substantial generalizations of Harris's strategy.

math.CA↗